1. Biochem. J. (1998) 334, 261–267 (Printed in Great Britain) 261
Amino acid availability regulates p70 S6 kinase and multiple
translation factors
Xuemin WANG*, Linda E. CAMPBELL*, Christa M. MILLER† and Christopher G. PROUD*1
*Department of Anatomy and Physiology, University of Dundee, Dundee DD1 4HN, U.K., and †Department of Biosciences, University of Kent at Canterbury,
Canterbury CT2 7NJ, U.K.
Incubation of Chinese hamster ovary cells without amino acids
for up to 60 min caused a rapid marked decrease in p70 S6 kinase
activity and increased binding of initiation factor eIF4E to its
inhibitory regulator protein 4E-BP1. This was associated with
dephosphorylation of 4E-BP1 and eIF4E and dissociation of
eIF4E from eIF4G. All these effects were rapidly reversed by
resupplying a mixture of amino acids and this was blocked by
rapamycin and by inhibitors of phosphatidylinositol 3-kinase,
implying a role for phosphatidylinositol 3-kinase in the signalling
pathway linking amino acids with the control of p70 S6 kinase
activity and the phosphorylation of these translation factors.
Amino acid withdrawal also led to changes in the phospho-
INTRODUCTION
Amino acid deficiency results in decreased rates of protein
synthesis in eukaryotic cells. One of the changes in the trans-
lational machinery which can be induced by amino acid depri-
vation is the increased phosphorylation of the α-subunit of
initiation factor-2 (eIF2α; eIF denotes eukaryotic initiation
factor), resulting in inhibition of the activity of the guanine
nucleotide-exchange factor, eIF2B, which is required to recycle
eIF2 and maintain high rates of translation initiation (reviewed
in [1–3]). This regulatory system has been most intensively
studied in yeast, which possess an eIF2α kinase, GCN2 (GCN,
general control, non-depressible), which is activated under con-
ditions of amino acid deficiency [4,5]. A GCN2 homologue has
recently been identified in Drosophila [6], although no such
enzyme has yet been reported from mammalian cells.
Other signalling pathways can also acutely regulate mRNA
translation in mammalian cells by affecting the states of phos-
phorylation and the activities of other translation factors or
regulatory proteins [2,7,8]. One such pathway is the rapa-
mycin-sensitive signalling pathway, which is activated by certain
hormones, growth factors and other mitogens, and is believed to
lead to the stimulation of a variety of regulatory steps in mRNA
translation. This pathway is also termed the FRAPmTOR
(FK506-binding protein–rapamycin-associated proteinmam-
malian target of rapamycin) pathway, referring to the protein
that is the target of rapamycin action. It is linked: (i) to the
regulation of the cap-binding translation factor, eIF4E, through
the regulatory proteins known as 4E-binding proteins (4E-BPs)
Abbreviations used: CHO, Chinese hamster ovary; eEF, eukaryotic elongation factor; eIF, eukaryotic initiation factor; eIF2α, α-subunit of initiation
factor-2; 4E-BP1, eukaryotic initiation factor eIF4E-binding protein-1; FRAP, FK506-binding protein-rapamycin-associated protein; GCN, general
control, non-depressible (GCN2 is a protein kinase acting on eIF2α); JNK, c-Jun N-terminal kinase; MAP, mitogen-activated protein; m7
GTP, 7-
methylGTP; Mnk1, MAP kinase signal-integrating kinase-1; mTOR, mammalian target of rapamycin; p70 S6 kinase, 70 kDa protein kinase acting on
ribosomal protein S6; PDK1, phosphoinositide-dependent kinase-1; PI 3-kinase, phosphatidylinositol 3-kinase; PKB, protein kinase B; 5h-TOP, 5h-
terminal tract of polypyrimidines.
1
To whom correspondence should be addressed (e-mail CGPROUD!bad.dundee.ac.uk).
rylation of other translation factors; phosphorylation of eIF4E
decreased whereas elongation factor eEF2 became more heavily
phosphorylated, each of these changes being associated with
decreased activity of the factor in question. Earlier studies have
suggested that protein kinase B (PKB) may act upstream of p70
S6 kinase. However, amino acids did not affect the activity of
PKB, indicating that amino acids activate p70 S6 kinase through
a pathway independent of this enzyme. Studies with individual
amino acids suggested that the effects on p70 S6 kinase activity
and translation-factor phosphorylation were independent of cell
swelling. The data show that amino acid supply regulates multiple
translation factors in mammalian cells.
or pH- and acid-stable (PHAS) proteins (reviewed in [9]); (ii) to
activation of p70 S6 kinase (the 70 kDa protein kinase acting
on ribosomal protein S6) [10], and hence the phosphorylation of
ribosomal protein S6, and, as suggested by recent data [11–13],
to the upregulation of the translation of mRNAs with 5h-
terminal polypyrimidine tracts (5h-TOP mRNAs); and (iii) to the
regulation of the phosphorylation of elongation factor eEF2 and
the control of elongation [14].
Here we demonstrate that amino acid availability regulates
this rapamycin-sensitive signalling pathway. Amino acid defi-
ciency results in the reversible inactivation of p70 S6 kinase. This
is associated with increased binding of the inhibitory protein 4E-
BP1 to eIF4E, and dissociation of the eIF4F complex, containing
eIF4G, which is involved in cap-dependent translation initiation.
Amino acid deprivation also brought about decreased phos-
phorylation of eIF4E and increased phosphorylation of eukaryo-
tic elongation factor eEF2. These data highlight new mechanisms
by which amino acid availability can regulate several translation
factors in mammalian cells.
MATERIALS AND METHODS
Chemicals and biochemicals
m(GTP (7-methylGTP)–Sepharose was from Pharmacia Biotech.
γ-$#P-labelled ATP, [$#P]Pi
and $&S-labelled methioninecysteine
were purchased from Amersham. Materials for tissue culture
were obtained from Gibco. Microcystin-LR and rapamycin were
from Calbiochem. The antiserum to rodent eIF4E has been

2. 262 X. Wang and others
described previously [15] and that to 4E-BP1 was raised against
a synthetic peptide corresponding to residues 101–113 of the
human protein. The antisera recognizing eIF4G and phosphory-
lated eIF2α were kindly provided by Dr. S. Morley (University
of Sussex, Brighton, U.K.) and Dr. G. Krause (Wayne State
University, Detroit, MI, U.S.A.) [16], respectively. Antisera to
individual isoforms of protein kinase B (PKB) were generously
provided by K. Walker and Dr. D. R. Alessi (University of
Dundee, Dundee, U.K.) [17].
Cell culture and treatment
Chinese hamster ovary (CHO.K1) cells were kindly provided by
Dr. L. Ellis (Houston, TX, U.S.A.), were maintained as described
previously [18] and were incubated without serum for 18 h prior
to use. To deprive the cells of amino acids, cells were transferred
to Earle’s balanced salt solution (Gibco-BRL). To restore amino
acids, an appropriate volume of a 10i stock of amino acids
(dissolved in Earle’s balanced salt solution) was added. This
stock contained tryptophan (0.6 mM), methionine (2 mM), his-
tidine (2.5 mM), tyrosine, cysteine and phenylalanine (all at
4 mM), arginine (5 mM), lysine (10 mM), threonine and the
branched-chain amino acids valine, isoleucine and leucine (each
at 8 mM), and glutamic acid (40 mM).
Assays for protein kinases, translation-factor phosphorylation and
association
The activity of p70 S6 kinase was measured after immuno-
precipitation using anti-(p70 S6 kinase) antibodies against a
peptide substrate as described previously [19]. PKB was assayed
(after immunoprecipitation with a mixture of antisera recognizing
all three isoforms of PKB [17]) using a synthetic peptide substrate
as described earlier [20]. The activities of p38 MAP (mitogen-
activated protein) kinase and c-Jun N-terminal kinase (JNK)
were assayed using heat shock protein 25 or a fusion protein
(glutathione S-transferase–c-Jun) as described previously [21,22].
MAP kinase activity was assessed by analysis of cell extracts in
Western blots using a commercial antibody (New England
Biolabs), which specifically recognizes the active phosphorylated
form of MAP kinase (both isoforms).
In order to assess the state of phosphorylation of 4E-BP1 by
radiolabelling, cells were preincubated with [$#P]Pi
for 150 min
in medium free of phosphate, and were then subjected to appro-
priate treatments. This was followed by extraction of the cells
in standard extraction buffer and immunoprecipitation of the
4E-BP1 [23], which was analysed by SDSPAGE and auto-
radiography.
eIF4E and associated proteins were isolated from cell extracts
by affinity chromatography on m(GTP–Sepharose and subjected
to SDSPAGE and Western blotting as described previously
[23,24] (any minor modifications are noted in the text). For
analysis of eIF4G, samples were run on an 8% polyacrylamide
gel and blots were developed with anti-eIF4G antiserum. eIF2α
phosphorylation was assessed by Western blotting using an
antibody to the phosphorylated form of the protein (to ascertain
the level of the phosphoprotein; [16]), which was then compared
with the total level of eIF2α in each sample, assessed using a
monoclonal antibody against eIF2α (a kind gift from the late Dr.
E. Henshaw, Rochester, NY, U.S.A.). The states of phospho-
rylation of eIF4E and of eEF2 were assessed by isoelectric
focusingimmunoblotting as described previously [14,25,26].
Immunoblots were developed by enhanced chemiluminescence.
RESULTS
Amino acid supply regulates the p70 S6 kinase
Removal of amino acids from confluent CHO cells resulted in
rapid inactivation of p70 S6 kinase, and this was reversed by
resupplying amino acids (Table 1a). The activation of p70 S6
kinase induced by restoring amino acids was completely blocked
by the immunosuppressant rapamycin (used at 25 nM) and by
either of two selective inhibitors of phosphatidylinositol 3-kinase
(PI 3-kinase), LY294002 (used at 30 µM) and wortmannin (used
at 100 nM; Table 1b). This is in line with data for its regulation
by other stimuli: rapamycin blocks p70 S6 kinase activation in
response to all agonists so far tested (for review, see [10]), and PI
3-kinase appears to be required for its activation by a wide range
of stimuli, but not by all of them [28–34]. Rapamycin and
wortmannin each repressed the basal activity of p70 S6 kinase in
serum-starved cells: cells were incubated for 60 min in serum-free
medium in the presence of rapamycin or wortmannin at the
standard concentrations. In rapamycin-treated cells p70 S6 kinase
activity was 26p4.5% (n l 6) of the serum-starved control,
whereas after wortmannin treatment the value was 24.8p5.9%
(n l 4), each being very similar to the 23% residual activity seen
after 60 min of amino acid withdrawal (Table 1a).
Table 1 Amino acid supply regulates p70 S6 kinase
(a) CHO.K1 cells were untreated (control), deprived of amino acids for the indicated times
or deprived of amino acids for 60 min followed by incubation in medium containing the
complete amino acid mixture for the times shown. Extracts were subjected to immunoprecipitation
with anti-(p70 S6 kinase) antiserum and the immunoprecipitates were assayed for p70 S6
kinase activity (see the Materials and methods section). (b) CHO.K1 cells were untreated or
deprived of amino acids for 45 min followed, where indicated, by incubation in the presence
of the full amino acid mixture with further additions of wortmannin (100 nM), LY294002
(30 µM) or rapamycin (25 µM).
(a) Time course of removal or resupply of amino acids
Condition/time
p70 S6 kinase
activity (% of control
pS.E.M.; n l 3)
Control (serum-starved) 100
Amino acid withdrawal
5 min 75.7p3.4
15 min 61.3p1.1
30 min 47.2p3.4
45 min 36.8p1.2
60 min 23.0p1.5
Amino acid resupply
5 min 69.1p3.5
15 min 110.0p8.0
30 min 105.9p9.2
(b) Effects of inhibitors on re-activation upon resupply of amino acids
Condition
p70 S6 kinase
activity (% of control
pS.E.M.; n l 3)
Control (serum-starved) 100
Amino acid withdrawal 41.5p4.5
Amino acids resupplied 126.2p20.0
jwortmannin 41.9p6.4
jLY294002 43.0p7.1
jrapamycin 40.4p3.9

3. 263Regulation of translation factors by amino acid availability
Figure 1 Amino acid withdrawal does not induce activation of p38 MAP
kinase or JNK
Serum-starved CHO.K1 cells were extracted without further treatment (lanes 2 and 3), pretreated
with 0.25 mM sodium arsenite for 20 min (lane 1) or subjected to amino acid withdrawal for
30 or 60 min (lanes 4 and 5) followed by resupply of amino acids for 30 min (lane 6) prior
to extraction. Samples were then analysed for p38 MAP kinase activity (using heat-shock protein
25 [hsp25] as substrate [21], A) or JNK (using a glutathione S-transferase (GST)–c-Jun fusion
protein as substrate [22], B). Samples were analysed by SDS/PAGE and autoradiography and
the Figure shows the resulting autoradiograph with the positions of the substrate proteins
indicated.
Table 2 Amino acid supply does not affect the activity of PKB
PKB was measured as described in the Materials and methods section (using a peptide
substrate after immunoprecipitation). CHO.K1 cells were either serum-starved or transferred to
Earle’s balanced salt solution (EBSS) for 60 min and, in some cases, resupplied with amino
acids for the times indicated. As a positive control for the activation of PKB, some serum-starved
cells were stimulated with serum (20%, v/v) for 30 min.
Condition/time
PKB activity (% of control
pS.E.M.; n l 3)
Control (serum-starved) 100
EBSS, 60 min 80.7p11.6
jamino acids for 5 min 59.3p4.8
jamino acids for 10 min 62.0p6.4
jamino acids for 20 min 63.2p7.3
Serum-starved plus serum 435p41.2
Amino acid withdrawal did not result in stimulation of the
activities of the stress-activated kinases, p38 MAP kinase or JNK
(Figure 1). If anything, a slight reduction was seen, although
since the basal activity was extremely low compared with that
elicited by treatment of the cells with arsenite, this change in
activity was very small indeed. Amino acid withdrawal had no
effect on the activity of ERK (the classical MAP kinase; results
not shown). It does not, therefore, elicit a stress response in terms
of activation of the stress-activated-kinase cascades over the time
period studied here, and does not affect the signalling pathways
upstream of the eIF4E kinase, MAP kinase signal-integrating
kinase-1 (Mnk1; p38 MAP kinase or ERK [35]).
PKB has been reported to lie upstream of p70 S6 kinase
[36,37]. Since it is regulated in a PI 3-kinase-dependent manner,
it has been suggested that it may provide a link between PI 3-
kinase and the activation of p70 S6 kinase. We therefore assessed
whether amino acids affect PKB activity in parallel with the
activation of p70 S6 kinase activity. Removal of amino acids did
cause a small fall in PKB activity (much less than the change in
p70 S6 kinase activity), but there was no increase in PKB activity
when amino acids were added back, although marked activation
was seen when CHO.K1 cells were stimulated with serum
(Table 2).
Amino acid deficiency leads to increased binding of 4E-BP1 to
eIF4E
The signalling pathway that regulates p70 S6 kinase also regulates
the phosphorylation of 4E-BP1 (reviewed in [9]), although there
appears to be a bifurcation in the pathway upstream of p70 S6
kinase and 4E-BP1 [38]. As shown in Figure 2(A), depriving
cells of all amino acids caused a rapid and marked increase
in the amount of 4E-BP1 recovered with eIF4E when cell
extracts were subjected to affinity chromatography on m(GTP–
Sepharose. This increase in 4E-BP1 binding was accompanied by
a decrease in the amount of eIF4G bound to eIF4E (Figure 2B),
as expected from the competitive nature of the interactions of
4E-BP1 and eIF4G with eIF4E [39,40]. The increased binding
of 4E-BP1 to eIF4E appeared to be due, again as expected, to a
decrease in its overall level of phosphorylation, as judged by its
increased mobility on SDSPAGE (Figure 2C). To confirm this,
the phosphorylation state of 4E-BP1 was assessed by radio-
labelling. As shown in Figure 3, incubation of cells in mixture
lacking amino acids resulted in decreased labelling of 4E-BP1.
This was reversed rapidly by addition of the amino acid mixture
to the cells. (Only two bands are visible on the autoradiograph,
as opposed to three on SDSPAGE, presumably because the
fastest-migrating one was not phosphorylated: this is similar to
earlier data from fat cells [23].) Thus the changes in phos-
phorylation status of 4E-BP1 inferred from its mobility on
SDSPAGE are borne out by examination by radiolabelling in
intact cells.
Resupplying the cells with amino acids resulted in a rapid
release of 4E-BP1 from eIF4E, and this was linked to increased
phosphorylation of 4E-BP1 and formation of eIF4E–4G com-
plexes (Figures 2A–2C). When tested across a range of dilutions
of the amino acid mixture, a 4-fold lower concentration of amino
acids caused substantial phosphorylation of 4E-BP1 and the
effect was essentially complete at half the standard concentration
(Figure 2D). Corresponding changes in the association of eIF4E
with 4E-BP1 and eIF4G were observed (results not shown).
Rapamycin and inhibitors of PI 3-kinase blocked the ability of
amino acids to induce release of 4E-BP1 from eIF4E and the
binding of eIF4G to eIF4E (Figures 2E–2G).
Amino acid supply regulates the phosphorylation of certain other
translation factors
eIF4E phosphorylation was assessed by isoelectric focusing
Western blotting. Amino acid depletion caused a modest but
reproducible decrease (typically from 30% in control cells to
15% without amino acids; four separate experiments) in the
phosphorylation of eIF4E, as manifested by an increase in
the intensity of the lower band (non-phosphorylated eIF4E)
relative to the upper band (phosphorylated eIF4E). This effect
was rapidly reversed by resupplying amino acids (Figure 4A).
Since the association of eIF4E with 4E-BP1 blocks the phos-

4. 264 X. Wang and others
Figure 2 Amino acid supply modulates the association of eIF4E with 4E-
BP1 and with eIF4G
(A–C) CHO.K1 cells were untreated (C), or incubated without amino acids for the indicated times
(Withdrawal of AA) or were incubated without amino acids for 45 min followed by resupply of
amino acids for the times shown (Resupply). Extracts were then prepared: for (A) and (B)
extracts were subjected to affinity chromatography on m7
GTP–Sepharose followed by SDS/PAGE
and immunoblotting using anti-eIF4E and -4E-BP1 antisera (A) or anti-eIF4G antiserum (B). For
(C), extracts were subjected directly to SDS/PAGE followed by Western blotting with anti-4E-
BP1 antiserum. (D) CHO.K1 cells were either untreated (C) or incubated without amino acids
for 45 min (k), followed by resupply of the standard amino acid mixture at the indicated
Figure 3 Amino acid supply regulates the phosphorylation of 4E-BP1
CHO cells were preincubated in the presence of [32
P]Pi (0.5 mCi/10 cm plate) in medium
lacking phosphate. Cells were then further incubated in medium containing (30 min, lane 1) or
lacking (15 min, lane 2; or 30 min, lanes 3–5) amino acids. After 30 min, amino acids were
added back to the cells for 10 (lane 4) or 15 (lane 5) min. 4E-BP1 was isolated from the extracts
by immunoprecipitation and analysed by SDS/PAGE (20% acrylamide gel) followed by
autoradiography. The migration position of 4E-BP1 (arrow) was identified by reference to
appropriate marker proteins of known molecular mass.
phorylation of eIF4E by Mnk1 [41], it is likely that the fall in
eIF4E phosphorylation was due to its increased binding to
eIF4E in response to amino acid depletion, resulting in inhibition
of its phosphorylation and consequent net dephosphorylation.
Consistent with this, the ability of amino acid restoration to
increase eIF4E phosphorylation was blocked either by rapamycin
or by inhibitors of PI 3-kinase, all of which also block the ability
of amino acids to induce dissociation of 4E-BP1 from eIF4E
(Figure 4A; cf. Figure 2E).
Elongation factor eEF2 is also subject to phosphorylation, the
phosphorylated form of the protein being inactive in translation
[42,43]. As shown in Figure 4(B), amino acid depletion resulted
in increased phosphorylation of eEF2. This effect was marked
after 45 min of amino acid withdrawal and the factor was almost
completely in its phosphorylated form after 60 min.
eIF2 and eIF2B
In other types of mammalian cells amino acid depletion has been
shown to cause increased phosphorylation of eIF2α [44]. How-
ever, in CHO cells, amino acid depletion did not bring about a
detectable change in eIF2α phosphorylation, even after 1 h
(Figure 4C) and, consistent with this, there was little if any
decrease in the activity of eIF2B (after 45 min of amino acid
withdrawal, eIF2B activity was 94.8p8.2% of control, n l 9).
Analysis of roles of individual amino acids
Addition of any of the amino acids in our standard mixture alone
(at the concentrations present in that mixture) did not affect p70
S6 kinase activity, the phosphorylation of 4E-BP1 or its binding
to eIF4E, or the association of eIF4E with eIF4G (Figure 4 and
results not shown). Given the very large number of possible
concentrations relative to the standard concentration (l 1.0) for 20 min. Analysis was as for
(C). (E–G) CHO.K1 cells were untreated or deprived of amino acids (E, for 45 min, k) for
the times shown (F, G) followed, where indicated (j), by incubation for 20 min in the presence
of the full amino acid mixture with wortmannin (W, 100 nM), LY294002 (LY, 30 µM) or
rapamycin (R, at 10 or 25 nM as indicated). (These inhibitors were added when amino acids
were withdrawn, and were therefore present for 45 min before amino acids were resupplied.)
Analysis for (E–G) was as for (A–C), respectively. The positions of eIF4E and 4E-BP1 are
indicated in (A) and (E): the major species of 4E-BP1 associated with eIF4E is α (least
phosphorylated). The positions of the α, β and γ species of 4E-BP1 are indicated in (C), (D)
and (G). eIF4G is indicated in (B) and (F).

5. 265Regulation of translation factors by amino acid availability
Figure 4 Amino acid supply modulates the phosphorylation of certain other
translation factors
(A) CHO.K1 cells were untreated (C) or incubated without amino acids for the times shown (min;
k30, k60). In some cases, after 60 min of incubation without amino acids, amino acids were
added (all lanes marked ‘j’) and the incubation continued for 5 or 10 min (j5, j10) or
for 20 min (j). In some cases, rapamycin (jR, 25 nM) or wortmannin (jW, 100 nM) were
added when amino acids were withdrawn, and were present when amino acids were resupplied.
Extracts were then subjected to affinity chromatography on m7
GTP–Sepharose followed by
isoelectric focusing analysis and immunoblotting using anti-eIF4E antiserum. The positions of
the unphosphorylated and phosphorylated forms of eIF4E are shown. (B) CHO.K1 cells were
untreated (C) or incubated without amino acids for the times shown. Extracts were subjected
to isoelectric focusing analysis followed by immunoblotting with anti-eEF2 antiserum. The
positions of the unphosphorylated and phosphorylated forms of eEF2 are shown. (C) CHO.K1
cells were untreated (C) or incubated without amino acids for 45 min (kAA). Amino acids were
then resupplied for 20 min (jAA). Extracts were subjected to SDS/PAGE and Western blotting
using either an antiserum specifically recognizing the phosphorylated form of eIF2α [eIF2α(P),
upper blot], or a monoclonal antibody recognizing eIF2α irrespective of its state of
phosphorylation (lower blot).
permutations, we have not tested combinations of two or more
amino acids. When added at higher concentrations (up to 5 times
the standard concentration), leucine alone could activate p70 S6
Figure 5 Effects of individual amino acids alone
CHO.K1 cells were untreated (Co) or incubated without amino acids (kaa) for 45 min. Where indicated, cells were then supplied with the full amino acid mixture (jaa) or individual amino acids
at the concentration found in the mixture; cysteine (C), histidine (H), isoleucine (I), L (leucine, also used at 4 mM, L4), methionine (M), threonine (T), tryptophan (W) or tyrosine (Y), in each case
for 20 min. Cell extracts were prepared and subjected to SDS/PAGE followed by Western blotting with anti-4E-BP1 antiserum. The positions of the α, β and γ species of 4E-BP1 are indicated.
kinase (to 78% of the control, the value for amino acid-depleted
cells being 46%, a mean of 2 experiments) and led to partial
phosphorylation of 4E-BP1 (Figure 4). The fact that leucine
alone is able to do this suggests that this phenomenon is not
associated with the cell swelling linked to Na+ ions, which leads
to activation of p70 S6 kinase in liver cells [45], since leucine is
not transported on a sodium-linked carrier in CHO cells [46].
Conversely, -aspartate, which is not a precursor for protein
synthesis but is transported on a sodium-linked carrier [46], did
not cause activation of p70 S6 kinase (data not shown), also
arguing against an effect mediated by cell swelling.
DISCUSSION
The data presented here demonstrate that amino acid availability
regulates the activity of the signalling pathway, which leads to
the activation of p70 S6 kinase. These data offer an explanation
for the observation [47] that amino acid deficiency results in
decreased phosphorylation of ribosomal protein S6 (which leads
to activation of autophagy, increasing the supply of intracellular
amino acids). Furthermore, they demonstrate that amino acid
supply modulates several important regulatory translation
factors through a variety of mechanisms. These include the
following.
(1) The availability of eIF4E, which is involved in cap-depen-
dent translation initiation and the modulation of several impor-
tant regulatory steps in translation, and forms a complex with
eIF4G and the helicase eIF4A (termed eIF4F [2,9,48]). By
bringing about the dephosphorylation of 4E-BP1, and thus
increasing its binding of eIF4E, amino acid deficiency results in
dissociation of the eIF4F complex. This is expected to lead to
decreased translation, especially of mRNAs whose 5h-untrans-
lated regions are rich in secondary structure, which inhibits their
translation, and which is believed to be unwound by eIF4A as
part of eIF4F.
(2) The phosphorylation of eIF4E, which decreases in response
to amino acid deprivation. Since phosphorylation of eIF4E en-
hances the affinity of eIF4E for mRNA [49] and may facilitate its
incorporation into initiation complexes [50,51], this effect could
contribute to the inactivation of mRNA translation too.
(3) The assembly of eIF4F. Amino acid resupply leads to for-
mation of eIF4E–4G complexes, which is probably linked to the
dissociation of eIF4E from 4E-BP1 and, perhaps, to the increased

6. 266 X. Wang and others
phosphorylation of eIF4E, which are induced by resupplying
amino acids.
(4) The phosphorylation of eEF2, which is increased by
removal of amino acids. Since phosphorylated eEF2 is inactive
in translation [42,43], this effect would result in inhibition of
elongation in parallel with the inhibition of initiation caused by
the preceding two effects.
The data also suggest that amino acid supply regulates the
translation of the 5h-TOP mRNAs. Although we have not
examined this in our present study, there is now substantial
evidence that the translation of these mRNAs is regulated
through the rapamycin-sensitive signalling pathway that leads to
p70 S6 kinase activation and phosphorylation of S6, which lies
in the mRNA-binding site of the 40 S subunit ([13]; reviewed in
[52]). These mRNAs encode proteins such as elongation factors
and ribosomal proteins. It makes excellent physiological sense
that the translation of such mRNAs should be shut off in
response to amino acid deficiency. p70 S6 kinase is also involved
in regulating cell-cycle progression [10,53,54] and the results
reported here point to a mechanism through which nutrient
(amino acid) availability could modulate cell growth and division.
Resupplying amino acids results in activation of p70 S6 kinase
and the reversal of the changes described above. The ability of
rapamycin to block this is consistent with the key role of RAFT
(rapamycin and FK506-binding protein-12 target)mTOR in the
activation of p70 S6 kinase in response to all known stimuli in
mammalian cells [10]. The fact that these effects were also
blocked by either of two structurally unrelated inhibitors of PI 3-
kinase, LY294002 and wortmannin, suggests that the regulation
of this pathway by amino acids involves one or more members of
the PI 3-kinase family of enzymes. It is currently unclear how
amino acid supply could be linked to the regulation of PI 3-
kinase(s). Earlier work has suggested that PKB, a protein kinase
that is regulated in a PI 3-kinase-dependent manner [20,55–58],
may lie upstream of p70 S6 kinase [36,37]. However, in our
experiments, amino acids had no effect on PKB activity under
conditions where they markedly activated p70 S6 kinase. Thus,
although amino acid-induced activation of p70 S6 kinase requires
PI 3-kinase, it seems to be independent of activation of PKB,
implying the operation of alternative upstream signalling path-
ways.
p70 S6 kinase has recently been reported to be a substrate for
FRAPmTOR [59] and for phosphoinositide-dependent kinase-
1 (PDK1) [60,61], a kinase that also phosphorylates and activates
PKB. The fact that rapamycin blocks the activation of p70 S6
kinase by amino acids indicates that FRAPmTOR is also
involved in this effect (as it is for all known activators of p70 S6
kinase in mammals). The activity of PDK1 against PKB is
constitutive, i.e. it is not enhanced, for example by insulin or by
3-phosphoinositides [60,62]. It is therefore unlikely that direct
regulation of this enzyme itself plays a role in the activation of
p70 S6 kinase (although its ability to phosphorylate p70 S6
kinase may well be influenced by the phosphorylation of other
sites in p70 S6 kinase [37]). We have not therefore measured its
activity in this study.
In certain other mammalian cell-types, and in Saccharomyces
cere isiae, amino acid deficiency has been shown to cause
increased phosphorylation of eIF2α andor inhibition of the
guanine-nucleotide exchange factor eIF2B (reviewed in [44,63]),
although this does not seem to be the case in the present
experiments. This may reflect the way in which the experiments
were performed. Earlier work, which showed increased eIF2α
phosphorylation, generally employed protocols that were likely
to lead to the accumulation of uncharged tRNAs (i.e. tempera-
ture-sensitive amino acyl-tRNA synthetase mutants [64,65], or
amino acid analogues that inhibit these enzymes [63]), whereas
this work involved removal of extracellular amino acids instead,
which may not cause accumulation of uncharged tRNA on the
time scale employed here. It is possible that uncharged tRNAs
activate a mammalian homologue of the eIF2α kinase GCN2,
leading to increased eIF2α phosphorylation, whereas removal of
external amino acids does not.
Taken together with the earlier work on eIF2 and eIF2B, it
seems that amino acids can regulate multiple translation factors
through the operation of at least two types of regulatory event
(the p70 S6 kinase or FRAPmTOR pathway and the phospho-
rylation of eIF2α), although the relative contribution of different
effects seems likely to vary depending, for example, on the
conditions. Both may play roles in the overall regulation of
translation (through the phosphorylation of eEF2 and eIF2α),
while the rapamycin-sensitive regulation of p70 S6 kinase and
of eIF4F is likely to be important in controlling the translation of
specific mRNAs (e.g. 5h-TOP mRNAs and those with extensive
secondary structure in their 5h-untranslated regions).
Since conducting our studies, Fox et al. [66] have reported that
amino acids stimulate the phosphorylation of p70 S6 kinase in
rat adipocytes and Xu et al. [67] have found that amino acids
induce partial phosphorylation of 4E-BP1 in islets of Langerhans.
This work was supported by a Programme Grant from the Wellcome Trust (to C.G.P.).
The initial observation that amino acid deficiency results in binding of 4E-BP1 to
eIF4E was made by R. Vries (University of Leiden, The Netherlands) while working
in this laboratory. We thank L. Wang for valuable assistance with the p70 S6 kinase
assays and Dr. J. McGivan (University of Bristol, Bristol, U.K.) for helpful comments
and advice.
REFERENCES
1 Price, N. T. and Proud, C. G. (1994) Biochimie 76, 748–760
2 Pain, V. M. (1996) Eur. J. Biochem. 236, 747–771
3 Clemens, M. J. (1996) in Translational Control (Hershey, J. W. B., Mathews, M. B.
and Sonenberg, N., eds.), pp. 139–172, Cold Spring Harbor Press, Cold Spring
Harbor
4 Hinnebusch, A. G. (1994) Trends Biochem. Sci. 19, 409–414
5 Hinnebusch, A. G. (1997) J. Biol. Chem. 272, 21661–21664
6 Santoyo, J., Alcalde, J., Mendez, R., Pulido, D. and deHaro, C. (1997) J. Biol. Chem.
272, 12544–12550
7 Rhoads, R. E. (1993) J. Biol. Chem. 268, 3017–3020
8 Redpath, N. T. and Proud, C. G. (1994) Biochim. Biophys. Acta 1220, 147–162
9 Lawrence, J. C. and Abraham, R. T. (1997) Trends Biochem. Sci. 22, 345–349
10 Proud, C. G. (1996) Trends Biochem. Sci. 21, 181–185
11 Jefferies, H. B. J., Reinhard, G., Kozma, S. C. and Thomas, G. (1994) Proc. Natl.
Acad. Sci. U.S.A. 91, 4441–4445
12 Terada, N., Patel, H. R., Takase, K., Kohno, K., Nairn, A. C. and Gelfand, E. W.
(1994) Proc. Natl. Acad. Sci. U.S.A. 91, 11477–11481
13 Jefferies, H. B. J., Fumagalli, S., Dennis, P. B., Reinhard, C., Pearson, R. B. and
Thomas, G. (1997) EMBO J. 16, 3693–3704
14 Redpath, N. T., Foulstone, E. J. and Proud, C. G. (1996) EMBO J. 15, 2291–2297
15 Flynn, A. and Proud, C. G. (1996) Eur. J. Biochem. 236, 40–47
16 DeGracia, D. J., Sullivan, J. M., Neumar, R. W., Alousi, S. S., Hikade, K. R., Pittman,
J. E., White, B. C., Rafols, J. A. and Krause, G. S. (1997) J. Cereb. Blood Flow
Metab. 17, 1291–1302
17 Walker, K. S., Deak, M., Paterson, A., Hudson, K., Cohen, P. and Alessi, D. R. (1998)
Biochem. J. 331, 299–308
18 Dickens, M., Chin, J. E., Roth, R. A., Ellis, L., Denton, R. M. and Tavare!, J. M. (1992)
Biochem. J. 287, 201–209
19 Moule, S. K., Edgell, N. J., Welsh, G. I., Diggle, T. A., Foulstone, E. J., Heesom, K. J.,
Proud, C. G. and Denton, R. M. (1995) Biochem. J. 311, 595–601
20 Cross, D. A. E., Alessi, D. R., Cohen, P., Andjelkovich, M. and Hemmings, B. A.
(1995) Nature (London) 378, 785–789
21 Cuenda, A., Rouse, J., Doza, Y. N., Meier, R., Cohen, P., Gallagher, T. F., Young,
P. R. and Lee, J. C. (1995) FEBS Lett. 364, 229–233
22 Minden, A., Lin, A., Claret, F.-X., Abo, A. and Karin, M. (1995) Cell 81, 1147–1157
23 Diggle, T. A., Moule, S. K., Avison, M. B., Flynn, A., Foulstone, E. J., Proud, C. G.
and Denton, R. M. (1996) Biochem. J. 316, 447–453
24 Flynn, A. and Proud, C. G. (1996) FEBS Lett. 389, 162–166

7. 267Regulation of translation factors by amino acid availability
25 Flynn, A. and Proud, C. G. (1995) J. Biol. Chem. 270, 21684–21688
26 Redpath, N. T. (1992) Anal. Biochem. 202, 340–343
27 Reference deleted.
28 Chung, J., Grammer, T. C., Lemon, K. P., Kazlauskas, A. and Blenis, J. (1994) Nature
(London) 370, 71–75
29 Dahl, J., Freund, R., Blenis, J. and Benjamin, T. L. (1996) Mol. Cell. Biol. 16,
2728–2735
30 Wilson, M., Burt, A. R., Milligan, G. and Anderson, N. G. (1996) J. Biol. Chem. 271,
8537–8540
31 Petritsch, C., Woscholski, R., Edelmann, H. M. L., Parker, P. J. and Ballou, L. M.
(1995) Eur. J. Biochem. 230, 431–438
32 Cheatham, B., Vlahos, C. J., Cheatham, L., Wang, L., Blenis, J. and Kahn, C. R.
(1994) Mol. Cell. Biol. 14, 4902–4911
33 Weng, Q. P., Andrabi, K., Klippel, A., Kozlowski, M. T., Williams, L. T. and Avruch, J.
(1995) Proc. Natl. Acad. Sci. U.S.A. 92, 5744–5748
34 Weng, Q.-P., Andrabi, K., Kozlowski, M. T., Grove, J. R. and Avruch, J. (1995)
Mol. Cell. Biol. 15, 2333–2340
35 Wang, X., Flynn, A., Waskiewicz, A. J., Webb, B. L. J., Vries, R. G. J., Baines,
I. A., Cooper, J. and Proud, C. G. (1998) J. Biol. Chem. 273, 9373–9377
36 Burgering, B. M. T. and Coffer, P. J. (1995) Nature (London) 376, 599–602
37 Peterson, R. T. and Schreiber, S. L. (1998) Curr. Biol. 8, R248–R250
38 Von Manteuffel, S. R., Dennis, P. B., Pullen, N., Gingras, A.-C., Sonenberg, N. and
Thomas, G. (1997) Mol. Cell. Biol. 17, 5426–5436
39 Haghighat, A., Mader, S., Pause, A. and Sonenberg, N. (1995) EMBO J. 14,
5701–5709
40 Mader, S., Lee, H., Pause, A. and Sonenberg, N. (1995) Mol. Cell. Biol. 15,
4990–4997
41 Vries, R. G. J., Flynn, A., Patel, J. C., Wang, X., Denton, R. M. and Proud, C. G.
(1998) J. Biol. Chem. 272, 32779–32784
42 Carlberg, U., Nilsson, A. and Nygard, O. (1990) Eur. J. Biochem. 191, 639–645
43 Redpath, N. T., Price, N. T., Severinov, K. V. and Proud, C. G. (1993) Eur. J.
Biochem. 213, 689–699
44 Proud, C. G. (1992) Curr. Top. Cell Regul. 32, 243–369
45 Krause, U., Rider, M. H. and Hue, L. (1996) J. Biol. Chem. 271, 16668–16673
46 McGivan, J. D. (1994) Biochem. J. 299, 321–334
47 Blommaart, E. F. C., Luiken, J. J. F. P., Blommaart, P. J. E., Vanwoerkom, G. M. and
Meijer, A. J. (1995) J. Biol. Chem. 270, 2320–2326
Received 5 February 1998/27 May 1998; accepted 23 June 1998
48 Morley, S. J., Curtis, P. S. and Pain, V. M. (1997) RNA 3, 1085–1104
49 Minich, W. B., Balasta, M. L., Goss, D. J. and Rhoads, R. E. (1994) Proc. Natl. Acad.
Sci. U.S.A. 91, 7668–7672
50 Sonenberg, N. (1996) in Translational Control (Hershey, J. W. B., Mathews, M. B.
and Sonenberg, N., eds.), pp. 245–269, Cold Spring Harbor Press, Cold Spring
Harbor
51 Flynn, A. and Proud, C. G. (1996) Cancer Surv. 27, 293–310
52 Jefferies, H. B. J. and Thomas, G. (1996) in Translational Control (Hershey, J. W. B.,
Mathews, M. B. and Sonenberg, N., eds.), pp. 389–409, Cold Spring Harbor Press,
Cold Spring Harbor
53 Reinhard, C., Fernandez, A., Lamb, N. J. C. and Thomas, G. (1994) EMBO J. 13,
1557–1565
54 Lane, H. A., Fernandez, A., Lamb, N. J. C. and Thomas, G. (1993) Nature (London)
363, 170–172
55 Franke, T. F., Kaplan, D. R., Cantley, L. C. and Toker, A. (1997) Science 275,
665–668
56 Klippel, A., Kavanaugh, W. M., Pot, D. and Williams, L. T. (1997) Mol. Cell. Biol. 17,
338–344
57 Alessi, D. R., Andjelkovic, M., Caudwell, B., Cron, P., Morrice, N., Cohen, P. and
Hemmings, B. (1996) EMBO J. 15, 6541–6551
58 Cross, D. A. E., Watt, P. W., Shaw, M., van der Kaay, J., Downes, C. P., Holder, J. C.
and Cohen, P. (1997) FEBS Lett. 406, 211–215
59 Burnett, P. E., Barrow, R. K., Cohen, N. A., Snyder, S. H. and Sabatini, D. M. (1998)
Proc. Natl. Acad. Sci. U.S.A. 95, 1432–1437
60 Alessi, D. R., Kozlowski, M. T., Weng, Q.-P., Morrice, N. and Avruch, J. (1998)
Curr. Biol. 8, 69–81
61 Pullen, N., Dennis, P. B., Andjelkovic, M., Dufner, A., Kozma, S. C., Hemmings, B. A.
and Thomas, G. (1998) Science 279, 707–710
62 Alessi, D. R., Deak, M., Casamayor, A., Caudwell, F. B., Morrice, N. and Norman,
D. G. (1998) Curr. Biol. 7, 776–789
63 Kimball, S. R., Antonetti, D. A., Brawley, R. M. and Jefferson, L. S. (1991) J. Biol.
Chem. 266, 1969–1976
64 Clemens, M. J., Galpine, A., Austin, S. A., Panniers, R., Henshaw, E. C., Duncan, R.,
Hershey, J. W. and Pollard, J. W. (1987) J. Biol. Chem. 262, 767–771
65 Pollard, J. W., Galpine, A. R. and Clemens, M. J. (1989) Eur. J. Biochem. 182, 1–9
66 Fox, H. L., Kimball, S. R., Jefferson, L. S. and Lynch, C. J. (1998) Am. J. Physiol.
43, C206–C213
67 Xu, G., Marshall, C. A., Lin, T.-A., Kwon, G., Munivenkatappa, R. B., Hill, J. R.,
Lawrence, J. C. and McDaniel, M. L. (1998) J. Biol. Chem. 273, 4485–4491